exploit
/ˈɛkˌsplɔɪt/A piece of code that takes advantage of a security weakness to attack a system.
Example: Hackers used an exploit to access customer records.
C222 words · Proficient
Information Technology vocabulary — C2
Expert IT vocabulary for security and systems reliability, including an exploit, a zero-day vulnerability, a race condition or memory leak, and the chaos engineering used to test for them.
22 questions
Words in this set
22zero-day
Describes a software flaw that is exploited by attackers before the developer has a chance to fix it.
Example:
The attack used a previously unknown zero-day flaw.
obfuscation
/ˌɑbfəˈskeɪʃən/Deliberately making code difficult to read or understand, often to protect it or hide its purpose.
Example:
The malware used obfuscation to avoid detection.
race condition
A software bug that occurs when the outcome depends on the unpredictable timing of two processes.
Example:
A race condition caused the order to be charged twice.
memory leak
A software fault where a program fails to free space it no longer needs, eventually slowing or crashing the system.
Example:
A memory leak made the app slower the longer it ran.
orchestration
/ˌɔrkəˈstreɪʃən/The automated management and coordination of multiple software systems or containers working together.
Example:
Orchestration software restarts any container that crashes.
immutable
/ˌɪˈmjutəbəl/Describes data or infrastructure that cannot be changed once it is created, only replaced.
Example:
The system treats each deployment as an immutable version.
buffer overflow
A software bug where data spills past the memory space set aside for it.
Example:
The hacker exploited a buffer overflow to run malicious code.
man-in-the-middle
An attack where someone secretly intercepts and can alter communication between two parties.
Example:
Public wifi made the laptop vulnerable to a man-in-the-middle attack.
sharding
Splitting a large database into smaller, separate parts to improve performance.
Example:
Engineers used sharding to spread user data across multiple servers.
byzantine fault
A failure where a component gives conflicting information to others.
Example:
One byzantine fault is harder to handle than ten clean crashes.
split brain
A state where two halves of a cluster both believe they are in charge.
Example:
Split brain wrote to both databases and nobody noticed for a day.
quorum
/ˈkwɔrəm/The minimum number of nodes that must agree for a decision to stand.
Example:
Losing quorum froze the whole cluster, which is the correct behaviour.
backpressure
A signal telling a fast producer to slow down.
Example:
Without backpressure the queue grew until the memory ran out.
thundering herd
Many clients retrying at once and overwhelming a recovering system.
Example:
The thundering herd took the service down again ninety seconds later.
cache invalidation
Deciding when stored data has become out of date.
Example:
Cache invalidation is famously one of the two hard problems.
side-channel attack
Learning secrets from timing or power rather than from the data.
Example:
A side-channel attack read the key from how long the check took.
supply chain attack
Compromising software by attacking one of its dependencies.
Example:
The supply chain attack came in through a package nobody had audited.
chaos engineering
Deliberately breaking things in production to prove resilience.
Example:
Chaos engineering found three single points of failure in a week.
fuzzing
Feeding random input to a program to find crashes.
Example:
A weekend of fuzzing produced eleven genuine security bugs.
blast radius
How much is affected when one component fails.
Example:
Splitting the service cut the blast radius to one region.
toil
/tɔɪl/Manual repetitive work that ought to be automated.
Example:
Half the team's week was toil and none of it was in the plan.
